Quantitative Trading Strategy, Backtesting, and Performance Analysis Using Python: A Data-Driven Analysis

Abstract

Background: The Nepal Stock Exchange (NEPSE) remains dominated by traditional trading behaviors, often driven by intuition and Fear of Missing Out (FOMO), rather than data or models. Despite global progress in algorithmic strategies, quantitative methods remain underutilized in Nepal’s equity market

Objectives: This study aims to design, implement, and evaluate a rule-based, long-only quantitative trading strategy tailored for NEPSE, comparing its performance to the standard Buy & Hold approach. The goal is to assess whether systematic models can improve capital protection and risk-adjusted performance in inefficient market settings.

Design/methodology/approach: A Python-based backtesting framework is developed, combining momentum and mean-reversion indicators such as the Z-score, Relative Strength Index (RSI), and a 240-day moving average to generate buy signals. Realistic trading constraints are incorporated, including dynamic position sizing based on risk, trade limits, and cool down periods. Performance is benchmarked against Buy & Hold using financial metrics such as CAGR, Sharpe Ratio, Sortino Ratio, Maximum Drawdown, Win Rate, Profit Factor, and Recovery Factor. Monte Carlo simulations, return distribution analysis, and sensitivity checks are applied to ensure robustness.

Results: While the strategy does not significantly outperform Buy & Hold in terms of raw daily returns, it shows superior performance on risk-adjusted metrics. It achieves a higher Sharpe and Sortino ratio and significantly lower drawdowns, confirming better risk control and capital preservation.

Conclusions: The findings suggest that even simple, rule-based quantitative systems can offer tangible benefits in emerging markets, such asNEPSE by reducing behavioral noise and enhancing portfolio discipline. The study supports the broader adoption of systematic methods in local trading environments.

Keywords: Quantitative Trading, NEPSE, Risk Management, Algorithmic Strategy, Backtesting.